Modeling the Initiation and Progression of Human MLL-Rearranged Precursor B Cell Acute Lymphoblastic Leukemia in Mice

Rearrangements of the mixed lineage leukemia (MLL) or lysine methyltransferase 2A (KMT2A) gene, accounting for approximate 10% of all leukemia cases and over 70% of infant leukemia cases, cause malignant and aggressive leukemias which usually respond poorly to chemotherapeutic agents. MLL-rearranged (MLL-r) leukemias include acute lymphoblastic leukemia (ALL), acute myeloid leukemia (AML), and mixed lineage leukemias. More than 80 different partner genes have been identified to fuse with MLL part. MLL fusions with AF9 and ENL genes account for over 40% of all the MLL-r ALL cases especially in infants and pediatrics. Numerous efforts have been devoted to developing MLL-r ALL mouse models to recapitulate human MLL-r leukemias. However, all the MLL fusions-driven mouse models often produce AML or biased towards AML, rather than B-cell ALL (B-ALL).

Here, we developed MLL-r ALL mouse models driven by MLL-AF9 or MLL-ENL that faithfully recapitulate corresponding human MLL-r B-ALLs. Briefly, we isolated and cultured mouse bone marrow (BM)-derived normal precursor B (pre-B) cells and then transduced them with retroviral MLL-AF9 or MLL-ENL (Fig. 1A, upper panel), with a strategy similar to the establishment of BCR-ABL1 or NRAS^G12D B-ALL mouse models as previously reported. Additionally, we included MLL-AF9-transduced lineage negative (Lin^-) progenitor cells-driven AML mouse model as a negative control group, which has been widely used in basic research and preclinical AML studies (Fig. 1A, lower panel). Our colony formation assay data showed that either MLL-AF9- or MLL-ENL-transduced normal pre-B cells, but not the empty vector (EV)-transduced group, could stably form colonies in vitro (Fig. 1B). After the BM transplantation (BMT) of the colony-forming cells into immunocompromised NSG mice via intravenously injection, we found that the organs (e.g. spleens) of recipients transplanted by MLL-AF9- or MLL-ENL-transformed pre-B cells were significantly enlarged compared to those of the control mice transplanted with normal pre-B cells (Figs. 1C and 1D). Our subsequent Wright-Giemsa stain of peripheral blood, Hematoxylin and Eosin (H&E) and Immunohistochemistry (IHC) staining against murine CD19 and CD11b of recipients’ spleens confirmed the development of full-blown B-ALL by MLL-AF9- or MLL-ENL-pre-B cells, in contrast to the AML phenotype of MLL-AF9-lin^-cells-transplanted mice (Figs. 1E and 1F). The overall survival (OS) time analysis showed that transplanted MLL-AF9 or MLL-ENL-pre-B cells could cause full-blown B-ALL within 3-5 months, while BCR-ABL1 B-ALL and MLL-AF9 AML models developed leukemia within 1 and 1-3 months, respectively (Fig. 1G). We then utilized the mixed leukemic BM cells from 3 randomly selected primary BMT recipients as donor cells for the 2^nd BMT, and then used mixed leukemic blast cells from 3 randomly selected 2^nd BMT recipients as donor cells for the 3^rd BMT. Our results showed that either MLL-AF9 or MLL-ENL B-ALL cells could rapidly expand and cause full-blown B-ALL in the 2^nd and 3^rd BMT recipient mice, with high white blood cell (WBC) counts in peripheral blood and significantly enlarged spleens in the 2^nd and 3^rd BMT (Figs. 1H and 1I). The OS time was significantly shorter in the 2^nd or 3^rd BMT recipients than in the 1^st BMT recipients (Figs. 1J and 1K), implying the mouse models were transplantable and aggressive. RNA-sequencing (RNA-seq) data showed that most of the dysregulated genes were upregulated in 1^st generation of MLL fusions-transformed B-ALL or MLL-AF9-transformed B-ALL in 2^nd recipients, with a significant enrichment of “positive regulation of transcription by RNA polymerase II” pathway in the overlapped upregulated-gene sets (e.g. Hoxa9 and Meis1 etc.) (Figs. 1L-1O). Our qRT-PCR results further validated the upregulation of target genes in mice samples (Figs. 1P and 1Q).

In sum, our study established transplantable mouse models recapitulating human MLL-r B-ALL in mice. Both MLL-AF9 and MLL-ENL could transform normal pre-B cells and develop phenotypic and morphological B-ALL within 3 months in 1^st recipients and only 20-40 days in 2^nd or 3^rd recipients in vivo. Gene expression analysis showed that MLL-AF9- or MLL-ENL-driven leukemogenesis in mice recapitulated human MLL-r B-ALL disease very well. Our mouse models highlighted applicable strategies to understand molecular mechanisms and test/develop therapeutic agents for human MLL-r B-ALL.